Project Summary Endocrine therapy plays a preeminent role in the management of the majority (about two thirds) of women with breast cancer whose tumors have the target, the estrogen receptor (ER). A recent meta-analysis showed that aromatase inhibitors (AIs) were superior to tamoxifen as adjuvant therapy in early-stage disease, but despite this superiority, about one-fifth of women had recurrence by 10 years. In addition to variability in outcomes, there is also a marked variability in tolerance, which can adversely impact adherence to treatment. The mechanism of action of AIs is inhibition of aromatase, thereby suppressing estrogen synthesis and reducing the ligand for the ER. The assumption is that all AIs produce sufficient estrogen suppression. However, our Preliminary Data showed marked variation in estradiol and estrone levels before and while on treatment with the AI anastrozole. However, it remains unknown whether the degree of estrogen suppression, or how to best quantify it, is related to degree of clinical benefit of these agents. We also have Preliminary Data from GWAS, using germline DNA from patients receiving AIs, showing that the variability in AI-related adverse events and outcomes is related to variation in host (germline) genetics. Furthermore, our preliminary findings with SNPs and genes identified to be associated with estrogen suppression by AIs and breast events (recurrences) provide a strong rational to test the hypothesis that genetic variation plays an important role in AI response, and this effect might be through the regulation of estrogen suppression, the mechanism of AI action. Of particular interest is that our functional studies of these genetic variants and genes also showed a SNP- and individual AI-dependent regulation of the expression of the aromatase gene. This novel finding has potentially important implications for precision AI treatment. Therefore, in this current proposal, we propose to take advantage of the extensive resources and Preliminary Data we have obtained. These include three major multi-center clinical trials involving all three third-generation AIs (anastrozole, exemestane, letrozole) for which we already have genome-wide genotyping available: 1) our own M3 study of anastrozole alone with estrogen levels pre and on anastrozole and anastrozole and anastrozole metabolite concentrations, 2) the MA.27 trial, comparing anastrozole and exemestane, from which we have published two GWAS relating to adverse events, and 3) the PreFace, a single-arm letrozole trial. MA.27 and PreFace have clinical follow-up data as well as biospecimens before and on AI treatment that would allow us to determine if the degree of estrogen suppression correlates with clinical AI treatment outcomes and, with genotyping data, the common or AI specific SNPs associated with these two phenotypes. We will then test the SNPs related to estrogen suppression in a prospective trial. Of crucial importance, we will also perform functional studies of those SNPs to elucidate mechanisms by which they might affect estrogen levels and AI response. These findings would have direct implications for the majority of women with breast cancer and would contribute to precision endocrine therapy with the AIs.